Scott M. Wilhelm

BRAF Is a Therapeutic Target in Aggressive Thyroid Carcinoma

Purpose: Oncogenic conversion of BRAF occurs in ∼44% of papillary thyroid carcinomas and 24% of anaplastic thyroid carcinomas. In papillary thyroid carcinomas, this mutation is associated with an unfavorable clinicopathologic outcome. Our aim was to exploit BRAF as a potential therapeutic target for thyroid carcinoma. Experimental Design: We used RNA interference to evaluate the effect of BRAF knockdown in the human anaplastic thyroid carcinoma cell lines FRO and ARO carrying the BRAF V600E (V600EBRAF) mutation. We also exploited the effect of BAY 43-9006 [N-(3-trifluoromethyl-4-chlorophenyl)-N-(4-(2-methylcarbamoyl pyridin-4-yl)oxyphenyl)urea], a multikinase inhibitor able to inhibit RAF family kinases in a panel of six V600EBRAF-positive thyroid carcinoma cell lines and in nude mice bearing ARO cell xenografts. Statistical tests were two sided. Results: Knockdown of BRAF by small inhibitory duplex RNA, but not control small inhibitory duplex RNA, inhibited the mitogen-activated protein kinase signaling cascade and the growth of ARO and FRO cells (P < 0.0001). These effects were mimicked by thyroid carcinoma cell treatment with BAY 43-9006 (IC50 = 0.5-1 μmol/L; P < 0.0001), whereas the compound had negligible effects in normal thyrocytes. ARO cell tumor xenografts were significantly (P < 0.0001) smaller in nude mice treated with BAY 43-9006 than in control mice. This inhibition was associated with suppression of phospho–mitogen-activated protein kinase levels. Conclusions: BRAF provides signals crucial for proliferation of thyroid carcinoma cells spontaneously harboring the V600EBRAF mutation and, therefore, BRAF suppression might have therapeutic potential in V600EBRAF-positive thyroid cancer.

Abstract SY11-02: Detection of tumor-associated mutations in circulating DNA: clinical applications and experiences.

The detection of tumor-associated mutations is of paramount importance in the era of personalized medicine. Mutational testing is now a prerequisite for the use of some approved therapies (e.g., KRAS for cetuximab in colorectal cancer [CRC]; BRAF for vemurafenib in melanoma), and these clearly established correlations between tumor mutational status and drug response elevate the importance and urgency of evaluating such associations in clinical trials of investigational drugs. While archival primary tumor tissue is often used for mutational evaluation, such material has inherent limitations, which may be overcome by recent technological developments enabling the detection of tumor-associated mutations using plasma-derived DNA. For example, when a tumor tissue specimen is unavailable, use of plasma DNA would allow mutational status to be ascertained without the need for an invasive procedure to obtain a new tumor sample. In addition, it is now apparent that most patients treated with targeted therapies will eventually develop drug resistance, often via the acquisition of new tumor-associated mutations; these mutations may vary not only between patients but also between metastases within an individual patient. As such, the mutational status of an archival primary tumor specimen may not be relevant to guide the selection of subsequent therapies, and obtaining fresh tumor tissue from each metastasis that arises following the development of drug resistance is impractical. In such instances, mutational analysis of DNA derived from a real-time plasma sample obtained after the onset of drug resistance may offer advantages in terms of both availability and biological relevance, since new mutations acquired in response to a particular targeted therapy may be detectable in plasma DNA. Finally, mutational analysis of plasma DNA may be useful in clinical trials to evaluate potential correlations between mutational status and clinical outcome. For such exploratory analyses, the collection of archival tumor specimens from a high proportion of enrolled patients can be logistically and ethically difficult to achieve, not to mention of questionable relevance given that acquired mutations would not be detectable in these specimens. Collection of fresh tumor tissue samples at study entry would provide biologically relevant material, but can be problematic and costly to obtain in large, global clinical trials. Thus, the utility of plasma DNA for real-time mutational analysis in the clinical-trial setting offers several distinct advantages. Since DNA derived from both normal and tumor cells exists in the circulation, the detection of tumor-associated mutations in plasma DNA requires the ability to identify a relatively small number of mutant alleles among an excess of wild-type alleles. With the goal of identifying the most suitable technology for this purpose, we conducted a comparison of available methodologies and found that BEAMing technology (Beads, Emulsions, Amplification, and Magnetics) offered very sensitive detection of known tumor-associated mutations using plasma DNA, although this technique is not well suited for the discovery of previously unknown mutations. We have now used BEAMing to analyze more than 2,000 patient samples collected from oncology clinical trials, allowing us to evaluate a number of genes (e.g., KRAS, NRAS, HRAS, BRAF, PIK3CA, AKT1, EGFR, KIT, and PDGFRA) in different cancer types (e.g., CRC, gastrointestinal stromal tumors, hepatocellular carcinoma, non-small-cell lung cancer, and breast cancer). We have used BEAMing of plasma DNA both prospectively, to enroll patients into a phase I trial based on a molecular profile of interest, and retrospectively, to evaluate potential associations between mutational status and clinical outcome in phase II and III trials. In many of these trials, collection of both fresh plasma and archival tumor tissue from a subset of patients has enabled us to compare mutational status in patient-matched plasma and tumor samples. Our experiences with BEAMing of plasma DNA to determine tumor-associated mutational status will be discussed. Citation Format: Michael Jeffers, Chetan D. Lathia, Scott M. Wilhelm, Dimitris Voliotis, Dirk Laurent, Carol E. Pena. Detection of tumor-associated mutations in circulating DNA: clinical applications and experiences. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr SY11-02. doi:10.1158/1538-7445.AM2013-SY11-02

Abstract 2337: Regorafenib (BAY 73-4506): Anti-metastatic activity in a mouse model of colorectal cancer

Regorafenib is an oral multi-kinase inhibitor which has recently been reported to extend overall survival in a Phase III trial in patients with metastatic colorectal tumors who have failed standard therapies (CORRECT trial) when given with best supportive care (BSC) vs placebo plus BSC. It is also examined in a Phase III trial for the treatment of patients with metastatic and/or unresectable gastrointestinal stromal tumors who have failed imatinib and sunitinib therapy (GRID trial). Regorafenib deactivates tumors across three dimensions (angiogenesis, oncogenesis, and stromatogenesis) by inhibition of angiogenic receptor tyrosine kinases (RTK) (e.g., VEGFR1-3, TIE2), oncogenic (e.g., KIT, RET) and stromal RTKs (e.g., PDGFR, FGFR). In preclinical studies it was previously shown that regorafenib inhibits a broad spectrum of different tumor types including colorectal cancer (CRC) models. Regorafenib was efficacious as a single agent and in combination with irinotecan in CRC models and demonstrated anti-metastatic activities in preclinical models of breast cancer. VEGFR-3, a member of the VEGF receptor family, is thought to play an important role in tumor lymphangiogenesis and metastasis. Regorafenib potently inhibited (4-8 nM) VEGF-C stimulated VEGFR-3 receptor autophosphorylation in human dermal lymphatic endothelial cells (LECs) and inhibited VEGF-C stimulated LEC cell migration in a scratch assay at concentrations below 100 nM. In vivo regorafenib9s antimetastatic potential was examined in a syngeneic CRC liver metastasis model. In this model murine MC38 CRC cells were injected into the mouse spleen with following ablation of the organ. Significant survival of mice treated orally at a daily dose of 10 mg/kg regorafenib was observed compared to vehicle-treated control mice. Liver weights, which were used as read-out for the degree of metastases formation, was reduced in mice treated with regorafenib for thirteen days vs vehicle treated animals. This was accompanied by significantly reduced number of metastases observed outside the liver. Finally, the direct effects of regorafenib in CRC cell lines was evaluated in vitro in a panel of 25 different human CRC cell lines where it inhibited the proliferation of 19 cell lines with IC50 values between 1 and 10µM and no correlation with mutated KRAS or BRAF was observed. Effects of regorafenib on intracellular MAPK signalling in CRC cell lines was analyzed by western blotting. Regorafenib, which also inhibits Raf kinases, reduced pERK by 50% at concentrations of about 500-2000 nM in HT29, HCT116 and Colo-205 CRC cell lines. In summary, these results suggest that regorafenib may exert its anti-tumor activity in colorectal cancer by inhibition of multiple kinases across its three dimensions within the tumor environment. Additional analyses in particular of clinical tumor samples are necessary to more clearly understand regorafenib9s molecular mechanisms of action. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 2337. doi:1538-7445.AM2012-2337

Abstract A43: Detection of tumor‐associated KRAS mutations in the plasma and tumor tissue of colorectal cancer patients

The mutational evaluation of tumor‐derived genetic material is becoming increasingly important in the era of personalized medicine. DNA harvested from a variety of sources (e.g., tumor biopsies, plasma, serum, urine and stool) may be useful to determine the mutational status of a tumor. In addition, a variety of methods for performing mutational analyses are available. In the present study, DNA derived from tumor biopsies and plasma was evaluated for KRAS mutational status by two different methodologies: Sequenom and BEAMing. In an initial experiment, KRAS mutations were identified in 36/64 (56%) of colorectal tumors analyzed by the Sequenom method. Matched plasma samples corresponding to a subset of the patients with KRAS‐mutant tumors (n = 18) were subsequently evaluated for KRAS mutational status by Sequenom and BEAMing methodologies. Each of these methods detected the expected tumor‐associated KRAS mutation in approximately 70% of the plasma samples analyzed. An additional set of 38 plasma samples from late stage metastatic colorectal cancer patients was then evaluated by both Sequenom and BEAMing, which lead to the identification of KRAS mutations in 9/38 (24%) and 19/38 (50%) of the samples analyzed by the respective technologies. Biopsies from 8 of these 38 patients were available and evaluated for KRAS mutational status by Sequenom. In one case, a KRAS mutation was found in the tumor that had not been detected in the corresponding plasma sample, and in another case a KRAS mutation that had been detected in a plasma sample was not detected in the corresponding tumor. This study indicates that the Sequenom method can be successfully used to determine the KRAS mutational status in both tumor biopsies and plasma. The BEAMing method was also successfully used to determine KRAS mutational status in plasma, and appears to be more sensitive than the Sequenom method. Since the frequency of KRAS mutation detected in the plasma of 38 metastatic colorectal cancer patients using BEAMing (i.e., 50%) is similar to that reported in the literature for this patient population when using tumor biopsies for mutational analysis, the assessment of tumor‐associated mutations in plasma may represent a viable option in this patient population. However, since the KRAS mutational status in plasma and tumor do not always correspond, data obtained from these two sources should be independently evaluated in the clinical setting. Citation Information: Mol Cancer Ther 2009;8(12 Suppl):A43.